Divergence can be identified only after selecting comparable swings on the same time axis and defining confirmation in advance. With RSI, price is compared with the oscillator; with the MACD-Histogram, the word can refer to a different pair of series. Signal confirmation, the time information became available, and risk must be assessed separately.
First identify two comparable price extremes and the matching extremes in the chosen indicator. Compare highs with highs or lows with lows; a gap between arbitrary points is not a documented Divergence. Indicator choice, Timeframe, calculation method and pivot rule all belong to the tested signal definition. [1] [3]
StockCharts describes a classic bearish Divergence when price makes a higher reaction high while RSI makes a lower value at the corresponding high. RSI fails to confirm the new price high and may warn of weakening momentum. In a strong uptrend, however, bearish Divergence can recur without an immediate reversal; keep the warning separate from confirmation of a price change. [1]
Classic bullish Divergence is the opposite relationship: price makes a lower reaction low while RSI makes a higher value at the corresponding low. This shows that RSI does not confirm the new price low; it does not establish a rising trend. Require a separate price or indicator condition, defined in advance, before concluding that a reversal has begun. [1]
This definition compares price swings directly with RSI. StockCharts also describes divergence between MACD and its histogram, which may precede a MACD signal-line crossover. It would be misleading to call that internal divergence automatically the same signal as a price–RSI mismatch; always name both series being compared. [1] [2]
A confirmed reaction high or low may require future bars. TradingView shows that pivot functions identify an extreme only after a specified number of following bars; plotting it back on an earlier bar can hide the actual discovery time on a historical chart. Backtesting must start only when all compared extremes were available under the rules, not on the date of a marker drawn in retrospect. [3] [4]
Divergence alone sets neither entry, invalidation nor target. Entry can depend, for example, on a later Close beyond a predefined price level, a change in price structure or an indicator crossover; each rule changes the trade's timing and price. StockCharts warns that bearish Divergence is common in strong trends and distinguishes an early MACD-Histogram warning from subsequent crossover confirmation. [1] [2]
Fix the market, Timeframe, RSI or MACD settings, pivot pairing, confirmation condition and validity period before testing. Include failed and incomplete patterns, Slippage, fees, liquidity and out-of-sample data. TradingView notes that a strategy's historical result depends on simulated orders and available data; a hand-picked chart does not demonstrate a trading edge. [4] [5]
The direction and size of the difference between two RSI highs are not a formula for the next price move. Set target price, risk and invalidation independently of the picture, before evaluating results. Without the specified confirmation, Divergence merely describes two series; even after confirmation, it guarantees no profit. [1] [5]
Price makes a higher high while RSI makes a lower high
For the clearest picture, read this entry together with Relative Strength Index, MACD, Technical analysis, Trendline, Backtesting. The reverse links also lead from MACD.
01Does bearish Divergence mean an immediate sell?+
No. A higher price high and lower RSI high describe a momentum mismatch. StockCharts warns that such a pattern can recur in a strong trend; entry needs a predefined confirmation and risk plan.
02Why can a historical chart show Divergence too early?+
A confirmed pivot may be recognized only after later bars, while its marker is drawn back over the original extreme. A test must use the actual time information became available, not the marker's position on the completed chart.